The Linux Thread - The Autist's OS of Choice

@dec05eba and others, have you managed to actually get xnamespaces working as intended? I've been messing with it for a few days but keep hitting a known bug that causes X_ChangeWindowAttributes to throw a BadAccess error, e.g. with urxvt and firefox:
Bash: 
 ns-launch ns-browsers urxvt
urxvt: An X Error occurred, trying to continue after report.
urxvt: + Serial number of failed request:  54
urxvt: + X Error of failed request: BadAccess (attempt to access private resource denied)
urxvt: + Major opcode of failed request:  2
urxvt: + (which is X_ChangeWindowAttributes)

ns-launch ns-browsers firefox
[Parent 26664, Main Thread] ###!!! ASSERTION: The program 'Firefox' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadAccess (attempt to access private resource denied)'.
  (Details: serial 243 error_code 10 request_code 2 (core protocol) minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the GDK_SYNCHRONIZE environment
   variable to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.): 'glib assertion', file /tmp/guix-build-firefox-149.0.2.drv-0/firefox-149.0.2/toolkit/xre/nsSigHandlers.cpp:1

(Firefox:26664): Gdk-ERROR **: 00:28:30.014: The program 'Firefox' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadAccess (attempt to access private resource denied)'.
  (Details: serial 243 error_code 10 request_code 2 (core protocol) minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the GDK_SYNCHRONIZE environment
   variable to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)
zsh: trace trap  ns-launch ns-browsers firefox

Where ns-launch is a small bash wrapper script for launching namespaces:
Bash: 
NS="$1"; shift
XAUTH="$HOME/.local/share/xns/${NS}.xauth"
[ -f "$XAUTH" ] || { echo "ns-launch: no xauth for '$NS'" >&2; exit 1; }
exec env XAUTHORITY="$XAUTH" "$@"

And the ns-browsers namespace definition is
Bash: 
namespace ns-browsers root
  auth MIT-MAGIC-COOKIE-1 [COOKIE]
  allow mouse-motion
  allow shape
  allow xinput
  allow xkeyboard
This is actively being worked on by the guy who originally found the bug and Metux in a dedicated fork, just figured I might as well ask if someone here has conjured up some black magic workaround while I wait for the pr to merge. IIRC there was talk somewhere of SonicDE implementing namespaces by default but I can neither find that claim nor actual code to back it up anywhere.
 
1777003572818.png
I once made this device into a fairly "Decent" Linux machine and even cellphone a long time ago via a bunch of mods and recompiling the kernel from scratch so it could "daulboot" both android and a linux install with a working x11 server.
 
Ferryman said:
@dec05eba and others, have you managed to actually get xnamespaces working as intended? I've been messing with it for a few days but keep hitting a known bug that causes X_ChangeWindowAttributes to throw a BadAccess error, e.g. with urxvt and firefox:
Click to expand...
Nah sorry I haven't used xnamespaces before. My guess is that its a missing case not handled in the xlibre code so you would need to report it as an issue.
 
I think it's a hardware issue, but I'd like to see if I could get some help with it. I use Linux Mint and a while ago, my laptop's speaker stopped after restarting with the power button due to a game crashing and capturing all my inputs. My headphones still work. I've tried using Alsamixer, checking the boot menu, reinstalling Pulseaudio drivers and using hdajackretask to override the audio pin, but none worked. I also tried live booting from a USB, but the audio still didn't work, which probably confirms it's a hardware issue. Does anyone have suggestions? Thanks.
 
Weird question, I'm running Debian 13 and I had docker installed and everything was working fine. but then I installed KVM so that i could use a windows VM and now any docker container running on the host can no longer access the internet (but the container is still perfectly accessible) and i don't understand why. I was able to move them to the bridge network in the meantime which seems to work, but now dlna and local discovery doesn't work anymore (not super important tho)
 
GratsTheat783 said:
I think it's a hardware issue, but I'd like to see if I could get some help with it. I use Linux Mint and a while ago, my laptop's speaker stopped after restarting with the power button due to a game crashing and capturing all my inputs. My headphones still work. I've tried using Alsamixer, checking the boot menu, reinstalling Pulseaudio drivers and using hdajackretask to override the audio pin, but none worked. I also tried live booting from a USB, but the audio still didn't work, which probably confirms it's a hardware issue. Does anyone have suggestions? Thanks.
Click to expand...
If it is on a LiveCD it gotta be hardware.
Betonhaus said:
Weird question, I'm running Debian 13 and I had docker installed and everything was working fine. but then I installed KVM so that i could use a windows VM and now any docker container running on the host can no longer access the internet (but the container is still perfectly accessible) and i don't understand why. I was able to move them to the bridge network in the meantime which seems to work, but now dlna and local discovery doesn't work anymore (not super important tho)
Click to expand...
People have experience similar but the other way around. Looks like it is something to do with docker, iptables and the KVM.

acid_ra1n said:
>biggest linux discord on the platform
>had a bot with source code freely available
>bot had admin access to the entire server
>someone hijacked the token and nuked everything
the owner is a tranny that vibe codes everything looks like it finally bit him in the ass
Click to expand...
I vibe coded some stuff today. I had to tell opencoder not to create sql injection vunerabilities.

A lot of developers generally have a very poor grasp of secure coding, opsec etc. I had a lot of security training at a particular contract and it opened my eyes.
 
Which part of linux would you not want to touch with a 10 foot pole to the point that you would rather just reinstall instead of touching its configuration files? For me its the audio subsystem, if that thing breaks i consider just reinstalling.
 
General Zoider said:
Which part of linux would you not want to touch with a 10 foot pole to the point that you would rather just reinstall instead of touching its configuration files? For me its the audio subsystem, if that thing breaks i consider just reinstalling.
Click to expand...
swapping DEs is a reinstall for me. way too much stuff like power managers, redundant applications, shitloads of dependencies and it never quite works out or if it does it doesnt feel right
 
SchizoDaemon said:
People have experience similar but the other way around. Looks like it is something to do with docker, iptables and the KVM.

https://codealpha.net/Homelab/Fix-Networking-issues-with-Docker-and-KVM-QEMU-on-Ubuntu/ https://forums.rockylinux.org/t/docker-ce-breaks-kvm-bridged-networking/11806 https://bbs.archlinux.org/viewtopic.php?id=233727
Click to expand...
the weird thing is that it's not a complete break, the virtual machine can connect to the internet fine, and docker containers using the bridge network i created work fine, the problem is that any docker container set to use "host" cannot access the internet (but are still fully accessible either directly or through a proxy). And weirdly enough it seems like I have three ip addresses for my computer, with the interface having it's own even though only the bridge should have two. Though that looks like it's due to the workaround that makes it so my server network connection doesn't go down just because I added a GPU or a PCIe storage device, and although I can fix it it'll just be a temporary fix.

Right now all the important stuff works but there's definitely something screwy.
Not something to tackle at 11pm when I have appointments the next day.
 
acid_ra1n said:
No I try it every now and then and the parts people like are just overengineered German bloatware and the rest of it is half working traah
Click to expand...
I would like something that has snapper + btrfs and uses mostly native system packages. I've tried to like flatpaks but honestly I have stopped liking them after a bunch of issues.
 
You must log in or register to reply here.
Back
Top Bottom